INDEX

Explanations

terms related to fire and firefighting

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_slefr-ajt/2-res_slefr-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

UGE

-0.70

 Vide

-0.69

laus

-0.66

Birth

-0.64

é¾įå¥ĳå£«

-0.61

 Duke

-0.61

DonaldTrump

-0.60

basketball

-0.57

 Goldberg

-0.57

VP

-0.57

POSITIVE LOGITS

 exting

0.84

 brigade

0.77

 retard

0.74

 blaze

0.74

gencies

0.73

 raged

0.69

ilies

0.69

 wildfire

0.68

 brig

0.68

 extingu

0.67

Activations Density 0.043%

terms related to fire and firefighting

No Comments

No Known Activations